Identity, equality and participation: testing the dimensions of citizenship in Canada and Belgium.

Chastenay, Marie-Helene ; Page, Michel ; Phalet, Karen 等

ABSTRACT/RESUME

The main purpose of the present paper is to develop the scientific concept of citizenship. In this article, we present two recent research projects concerned with the simultaneous measure of the multiple dimensions of citizenship at the individual level and the identification of the relations and distinctions between them, one such study conducted on a sample of young Canadian citizens, the second on a Belgian sample composed of Belgian citizens and Turkish and Moroccan non-national residents of Belgium. Although their methodology differs considerably, both studies identified three core, theoretical dimensions that are remarkably similar and coincide with those discussed in related empirical studies and theoretical models: identity, social equality, and norms, and participation. Results reveal common tendencies, notably the possibility of a link between aspects of identity and participation, as well as variations suggesting that the relationship among variables is more complex than one might wish for in a citizenship model aspiring to universality. Although officials need to have access to a relatively clear definition of what citizenship encompasses, this implies a certain normalization of an ideal that might not reflect actual forms of citizenship present in the population. Such research can thus contribute to the development of public policies that relate the associated needs to forms of citizenship as these are experienced by ordinary citizens in their daily life.

Le principal objectif de cet article est le developpement du concept scientifique de citoyennete. Nous presenterons deux projets de recherche recents visant a mesurer simultanement les multiples dimensions de la citoyennete individuelle et a identifier les relations et distinctions entre ces dimensions, l'un realise aupres de jeunes citoyens canadiens et le second aupres d'un echantillon belge compose de citoyens belges et de Turcs et Marocains residant en Belgique. Bien que leurs methodologies respectives different considerablement, les deux etudes ont identifie trois dimensions principales qui sont remarquablement similaires et coincident avec celles qu'on retrouve dans d'autres etudes empiriques et modeles theoriques : identite, egalite et normes sociales, et participation. Les resultats revelent des tendances communes, notamment la possibilite d'une lien entre les aspects identitaires et participatifs, de meme que des variations qui suggerent que la relation entre les variables est plus complexe que dans les modeles universels de citoyennete. Bien que les decideurs doivent avoir acces a une definition claire de ce que le concept de citoyennete recouvre, ceci implique une certaine nonnalisation d'un iddal qui ne correspond peut-etre pas aux formes de citoyennete presentes au sein de la population. De telles recherches peuvent donc contribuer au developpement de politiques publiques qui refletent les besoins associes aux formes de citoyennete vecues par les citoyens ordinaires dans leur vie quotidienne.

INTRODUCTION

The main purpose of this paper is to develop the scientific concept of citizenship. Social psychologists are accustomed to multidimensional concepts, their scope and dimensions being precisely circumscribed by factor analysis and by research data that also show how these dimensions are correlated. Although these dimensions can be considered to measure different constructs, their covariation reveals their relationship.

Very few studies address the simultaneous measure of the multiple dimensions of citizenship at the individual level and the identification of the relations and distinctions between them. In this article, we present two such recent research projects, one conducted on a sample of young citizens in three Canadian provinces, the second on a Belgian sample composed of Belgian citizens and non-national Turkish and Moroccan residents of Belgium. We first review theoretical dimensions of citizenship and the comparability of the Canadian and Belgian contexts, and then present the two studies and their results. We conclude by discussing common trends emerging from the comparison, as well as variations found across the sub-samples and societies.

It should be noted that the two studies were undertaken and conducted independently and have already resulted in various distinct publications (Page and Chastenay 2002, 2003; Phalet and Swyngedouw 2001,2002; Swyngedouw et al. 1999). Although no international comparison was intended in their initial design, the present comparative article stems from the creation, in 1998, of the Network on Education in Divided Societies, composed of researchers from Belgium, Quebec, Northern Ireland, and Catalonia. It was more specifically initiated during the Quebec-Flanders seminar held in Ghent in 2002, where both groups of researchers met and discussed the similarities and differences of the studies and their potential comparability.

Although the main theoretical dimensions are surprisingly similar and both studies used a questionnaire, their methodology differs considerably (for example, in the choice and size of the sample, the type and number of questions, etc.), clearly limiting their comparability.' The reader should thus not be surprised that the methodology and results of the Canadian study are discussed at greater length than those of the Belgian study, given the preponderance of the Canadian data.

THEORETICAL DIMENSIONS OF CITIZENSHIP

The first section presents the dimensions of the concept of citizenship used in the development of the instruments in the studies. They both identified three core theoretical dimensions that are remarkably similar and coincide with those discussed in related empirical studies and theoretical models: identity, social equality and norms, and participation.

Citizenship Dimensions in the Canadian Study

The study conducted with young Canadians is based on a conceptual framework developed by Gagnon and Page (1999) that allows for an overall vision of citizenship as defined in a few liberal, democratic societies. Four dimensions emerged from this vast review of the relevant literature: national identity, cultural (group) identity, the relationship to equality norms, and the relationship to civic participation.

The national identity dimension is first defined by legal and political principles included in the constitution of each society. It is also defined by societal culture which refers to all characteristics of the individuals' public mode of living in the society: its system of production of goods, its national sports, the most common life habits, its particular architecture, and so on. Institutional norms that govern the functioning of the society's institutions include the official language(s) used by civil servants working in public institutions. Each society is also characterized by its media, which, in some cases, reflect the society's predominant ideological and political currents. Heritage is yet another sub-concept of national identity that can be divided into several elements relative to the natural environment, history (including symbols and founding myths, among which we find, for example, the historical development of democratic institutions and of the legal framework), as well as cultures, heritage languages, and cultural production (art works, architecture, etc.).

Allegiance and patriotism also constitute components of national identity, as their definition refers to the latter; patriotism is indeed defined in terms of attachment and loyalty to a country and its political community.

Social, cultural, and supranational membership is a second aspect of citizenship refering to the various forms of acknowledged diversity present in the society: national minorities who generally enjoy a distinct legal and political status, those with a strong regional identity, or yet other minorities, whether they be cultural, religious, or sociological (gays, senior citizens, women, youth, disabled, etc.). We must also take the supranational level into account, a form of identity that allows membership to a community that does not necessarily identify with a state or specific geographical location, but rather to supranational groups formed on the basis of common interests, religion, etc.

Citizenship is also incarnated in a system of laws comprising public norms that define the citizen's political and legal status: there are acknowledged citizens' rights (fundamental, political, social, and cultural) as well as programs and measures put in place to ensure equitable access to economical, societal, and institutional resources.

In a democracy, participation in political life and civil society presents citizens with both rights and responsibilities. Political participation includes exercising the right to vote during elections, becoming involved in political discussions, keeping informed, becoming a member of a political party, working in electoral campaigns, contacting politicians, running for election in public functions, and, finally, assuming such functions. Civil participation is conceived as volunteer involvement in organizations independent from the state, with activities that are entirely dependent on their members' initiative. Most such organizations are active on a local scale (parish, school, neighborhood, city), but others, such as the Red Cross, are active internationally. Participation in the management of state institutions also constitutes a form of civil participation that seems to be growing in importance: it is achieved through users' committees, governing boards, parent-teachers' associations, etc. An important aspect of this last dimension refers to the abilities citizens require in order for them to participate, their motivations and the information they need to allow efficient participation.

Citizenship Dimensions in the Belgian Study

Phalet and Swyngedouw (2001) present three core dimensions that served as the basis for the development of the questionnaire they used to analyze the social representations of transnational citizenship of a sample of Turks and Moroccans living in Belgium. One of the main objectives of their study was to establish whether a distinction could be empirically demonstrated between dimensions that are often confounded in the analysis of the citizen-state relation.

According to them, the identify, dimension first relates to membership in the national community, serving as a formal criterion defining the citizen-state relationship. This classical perspective on citizenship implies that forms of collective identity not recognized by the nation state (religious, regional, linguistic, ethnic, etc.) are also not endowed with political legitimacy, and thus should not be acknowledged in the public sphere. However, this view cannot adequately account for citizens' identities, which are both multiple and subjective (as is easily illustrated in the case of immigrants) and which are more or less tolerated in different political systems. The possibility for minority groups to have their claims heard and acknowledged (and the allocation of power that this implies) can thus vary greatly across different models of citizenship.

The normative dimension refers to socialization to norms and democratic or "civic" values, including notions such as rights, freedoms, duties, and responsibilities, all of which are often discussed in public debates about citizenship. However, theoretical normative models disagree as to which should be the primary source of this socialization (family, education, formal public participation, etc.). In this perspective, immigrants' experience could be a particularly fruitful research area for the analysis of multiple sources of influence and their combined effect on this socialization process.

Another principal dimension is participation, whereby citizens have access to public institutions and services. It also includes their potential involvement in formal or informal public (or civic) activities. Here again, theoretical nonnative models differ as to the degree and forms of participation expected of citizens. Whereas classical perspectives mainly focus on access, new models also consider informal associations, arguing that active participation is an essential component of a democratic system that should also include immigrant participation.

Compared with the Canadian study, the identity and participation dimensions have similar meanings. However, it should be noted that the Belgian normative dimension refers to common national values, whereas the Canadian equality dimension measures openness and acceptance of cultural pluralism, which mainly reflect multiculturalism and values related to diversity. In comparing factorial results, one should therefore take into account the fact that they might partly reflect different prior conceptualizations that can be attributed to differences in the national histories and integration paradigms of the two countries.

Citizenship Dimensions: Related Studies

It is interesting to note that the same dimensions of the concept of citizenship are found in other empirical studies that share an interest in the ways in which individuals conceive of citizenship in different countries.

Let us consider, as a significant example, the 2001 study directed by Torney-Purta et al. evaluating the impact of civic education in twenty-eight countries under the patronage of the International Association for the Assessment of Educational Achievement (IEA). The variables used in the survey correspond to objects of various psychological entities similar to those measured in the Canadian and Belgian studies (such as concepts, interests, intentions, self-reported behaviors, attitudes, and values).

The scale measuring the concept youths have of a democratic society comprises several items referring directly to citizens' rights in a democracy. Among the moderately or highly consensual items, we find, for example, the right to freely elect political leaders, women's right to be elected to government office, freedom of speech and of association, the social right to subsistence, the right to be informed by a free press, the right to preserve an ethnic identity, and so on.

The "good citizen" concept is measured by two scales referring to participatory behaviors in public affairs. One is composed of conventional political participatory activities, while the other covers social movements' activities. These two dimensions are also the object of a measure of the intention to get involved, presently and in the future, in the political life of the country, social movements or causes, charity, public protest, and/or illegal acts, such as blocking traffic or occupying a building. The relationship youths have with political participation outside of school is measured in this survey by scales of interest in political news in the media. Participation is also measured through activities practiced inside school and covering a wide range: learning democratic behaviors, trust in the efficiency of participation in the school, and participation in classroom discussions and/or student organizations in which they might be involved.

In addition, the survey covers attitudes and values toward different objects belonging to one of the three dimensions. For example, the participation dimension includes trust in the government, and the identity dimension is assessed through items relating to "national feeling and attachment to the country and its political symbols." These two scales address both support for the regime (trust and confidence in political institutions) and support for the political community (national pride). The rights dimension covers equality of rights for all citizens (native or new, man or woman) with two scales measuring attitudes and values. The first is "the extent to which students support certain rights or opportunities for immigrants," while the second probes "the extent to which they endorse political and economic rights for women" (p. 91).

Another author uses a similar conceptual framework of citizenship to present an integrated vision of the ways immigrant citizens exercise their status and the impact of their practices on the various conceptions of citizenship thus revealed (Bloemraad 2000). In order to carry out an analysis from this perspective, Bloemraad uses three dimensions that are very similar to those found in our studies; identity, rights, and participation.

Regarding rights as a dimension of citizenship, we know that, in most immigration countries, civil and social rights are not necessarily linked to citizenship, and that immigrants who do not have citizenship status also enjoy these rights. An important part of these rights are thus not linked to citizenship, at least in western countries that host large numbers of immigrants.

 According to Soysal, international migration, new political bodies
 such as the European Union, and the international discourse of
 human rights have created a situation where rights are now
 deterritorialized and located in the person rather than in an
 individual's nationality-based relationship to a state. (Bloemraad
 2000, 18)

Only the right to political, and not civic, participation is still closely linked to citizenship, but this restriction finds as many critics as defendants in the debate surrounding this aspect of citizenship.

Bloemraad follows a similar demonstration concerning identity. She reviews the everlasting debate between a liberal and a communitarian conception of citizenship, where the link uniting the citizen to the socio-political community is precisely the object of divergence. The liberal conception, insisting on the freedom of the individual citizen to pursue his/her personal objectives, is seen as a weakening of this link in as much as having the pursuit of personal goals as a priority implies keeping one's distance from the collective identity and the common interest. Conversely, the communitarian perspective insists on the necessity of this link. The multicultural conception of citizenship views adherence to the political community at large through adherence to a group identity recognized by this community. Although this debate suggests that all do not endorse the close link between citizenship and collective identity, studies conducted with new citizens in the United States and in Canada show that they nevertheless express a strong adherence to the host country's national identity

Participation is the third dimension analyzed by the author. It is a generally acknowledged fact that participation strengthens the bond between the citizen and the socio-political community. However, research in the field of immigrant participation shows that this bond is not as universal as we might think. Immigrants' interest in their country of origin and their participation in political affairs force us to rethink the relationship between citizenship and the nation state. "Transnationalism--occupations and activities that require regular and sustained social contexts over time across national borders for their implementation (Portes et al. 1999)--challenges traditional conceptions of citizenship" (Bloemraad 2000, 29). Obviously, maintaining political activity in the country of origin is a controversial subject in normative debates regarding citizenship.

However, when examining the relationship between citizenship and participation, one nmst consider more than participation beyond the host society's political community. Above all, we must not forget that there are barriers, obstacles, and constraints that can jeopardize the new citizen's access to participation. Knowing the extent to which acquired citizenship guarantees free access to all forms of participation is a principal research preoccupation, as there is consensus regarding the need to define the conditions that promote the active presence of new citizens in political activities and among elected representatives.

Citizenship Dimensions in Theoretical Normative Models

Even as the present paper discusses empirical studies, we must consider that these dimensions are also found in theoretical normative models that define different perspectives on citizenship, among which are the well-known liberal (Rawls 1993), republican (Schnapper 1994), and multicultural (Kymlicka 1995) models. It is possible to characterize each of these models using the dimensions identified above (Page 2001), as can be seen in the following contrast between two very different conceptions. The liberal model is characterized by a moderate identification to the national community and a strong identification to other groups chosen by the individual. This particular conception is thus characterized by a certain level of tolerance toward diversity. A strong adherence to a system of laws is also associated with this conception, the first foundation of which is the community's agreement with principles of justice. There is only moderate involvement of citizens in political participation, since they mainly participate when they feel a threat to their individual rights. Their participation in civil society is largely concentrated in associations involved in the protection and defense of individual rights and freedoms (Page 2001).

Alternately, the republican conception of a unitarian nation-state advocates strong identification with the national community that bears the identity of the majority group and weak identification with other groups, associated with a low tolerance for manifestations of non-national identifications. In such a context, the possibility for members of minority groups to enjoy equal status with members of the majority group is conditional upon their identification with the national community. Therefore, a distinct cultural life should, according to this conception, be restricted to private and intracommunity life. Citizens who adhere to this view show a strong disposition to engage in civic life, as it affords them the best opportunity to contribute to maintaining and enhancing national identity (Page 2001).

In the coming discussion, we rely on normative models to assist us in clarifying the interpretation of data. However, the two studies presented below did not attempt to validate such theoretical models, but rather to develop descriptive instruments for the empirical study of actual forms of citizenship.

Comparability of the Canadian and Belgian Contexts

By studying citizenship along these three dimensions, and more particularly how they are related or distinct, it is possible to reveal how constitutional and social frameworks of citizenship are experienced in Canada and Belgium, these being two countries that illustrate interesting comparative elements. Both are characterized by a strong, internal cultural diversity emerging from two sources, the first of which can be defined as follows:

 One source of cultural diversity is the coexistence within a given
 state of more than one nation, where "nation" means a historical
 community, more or less institutionally complete, occupying a given
 territory or homeland, sharing a distinct language and culture. A
 "nation" in this sociological sense is closely related to the idea
 of a "people" or a "culture".... [A country] which contains more
 than one nation is ... a multination state, and the smaller
 cultures form "national minorities." (Kymlicka 1995, 11-12)

However, although this definition of a multination state applies equally to Canada and Belgium, the two countries differ on the constitutional arrangements that establish the status of nations comprising each country. Belgium is a federal state with a constitutional monarchy composed of three communities (Flemish, Walloon, and Germanic) and three regions (Flemish, Walloon, and Brussels). Each community and region is endowed with autonomous governing structures exerting power in areas deemed essential to the maintenance of the various national identities. The federal state holds centralized powers in areas such as taxes, social security, justice, defense, and international relations.

Canada is an officially bilingual (French and English) and multicultural federal state (although there are also provincial policies governing language issues). The great majority of Francophones live in the province of Quebec which, though it does not enjoy a distinct constitutional status, is still considered the homeland of the Quebecois nation. Powers devolved upon provinces-notably in education, health, immigration, and the exploitation of natural resources are largely used in Quebec to preserve the Francophone national identity and the specificity of Quebec institutions. The remaining provinces are Anglophone with the exception of New Brunswick, which is officially bilingual, and they co-operate, between provinces and with the federal government, to develop institutions meant to incarnate the image of an Anglophone Canadian specificity. In Canada, the federal government exerts power in areas that ensure its strong presence throughout the country and, in the last few decades, it has been pursuing a policy to share jurisdiction with provinces, if not to centralize it. A governor-general acts as the official representative of the British crown.

A second source from which stems a country's internal cultural diversity is also the result of international migration. A country is multicultural when it welcomes significant numbers of foreign individuals and families and allows them to preserve their cultural particularities.

Canada has one of the highest immigration rates per capita, Immigrants can become citizens after three years of residence; they are free to maintain their customs if they wish to do so, to associate freely within their culturally distinct community, and to practice their religion. In this way, norms in a multicultural society allow them to experience and maintain their ethnocultural particularity, not only in their private life, but also within the host society's public institutions. With a federal multiculturalism policy, it ensues that one of the dominant political discourses advocates for a multicultural type of citizenship.

Ethnocultural groups in Canada do not possess powers of self-government. With a pluralistic conception of citizenship, their particular rights enable them "to express their cultural particularity and pride without it hampering their success in the economic and political institutions of the dominant society" (Kymlicka 1995, 31). In Quebec, this right to the preservation of ethnic culture is protected by the Charte des Droits el Libertes de la Personne du Quebec (Quebec Charter of Individual Rights and Freedoms). Similar rights are also guaranteed by the Canadian Charter of Rights and Freedoms.

In Canada, the Francophone and Anglophone national groups do not constitute an entity relating to a single common ancestry. Anglophone Canadians of Anglo-Saxon descent are now only a minority in the total Anglophone-Canadian national majority, resulting from a long history of immigrant integration. This is also true for the Quebec national Francophone minority, which is a mixed entity comprising a majority descended from French settlers, as well as an increasing number of new citizens with greatly varied ethnic origins.

Belgium shares its history of nation formation and immigration with continental Europe, and this shared history generated important differences from the Canadian case. Belgium has only recently started to develop policies with regard to ethnic cultural diversity. After the end of official immigration in 1974, it is still possible for foreigners to acquire Belgian nationality through policies of family reunion. A large number, however, have not yet requested the naturalization of their status and hence are designated by the formal nationality of their countries of origin, like Turkey and Morocco. These non-nationals enjoy the same social rights as Belgian citizens, but do not have access to political rights (with the exception of those granted to European citizens who may participate in municipal political affairs). The fact that non-national residents originating from outside Europe may only participate in the political affairs of their country of origin has led to a growing interest in transnational forms of citizenship.

Thus, the two multinational and multiethnic countries from which our samples are drawn are comparable in the way they generate constitutional arrangements taking national groups into account, as well as policies that define the rights of ethnocultural minorities stemming from immigration.

Testing the Dimensions of Citizenship

Our aim in the present paper is to demonstrate that studying the relations and distinctions between dimensions of citizenship reveals both commonalities and specificities across the two countries. Two main questions guided us through the process: "How do Canadian and Belgian national groups construct representations of their own citizenship (as described by the three dimensions mentioned in the introduction)?" and "How are new Canadian citizens and non-national residents in Belgium characterized by these same dimensions?" This study is thus descriptive, in as much as the results reveal how individuals combine our descriptive dimensions to define themselves as citizens.

THE CANADIAN STUDY

In this article, we locus on the results of an analysis that allow us to see how previously identified variables are related to each other to form the dimensions of the concept of citizenship. The crucial question is the following: Are these factors associated, and do they form the same dimensions found in Gagnon and Page's (1999) conceptual framework, or do they, rather, create other dimensions? We thus first specify the dimensional structure of the concept of citizenship and then identify the relationship between these dimensions.

Sample

This study was conducted with three samples from Quebec, New Brunswick, and Alberta. In all three provinces, the aim was to build a sample that could reflect the variety of contexts in which young adults pursue their post-secondary studies, not one that would be strictly representative of the population of this age bracket. Therefore, when possible, Francophone and Anglophone colleges and universities were selected in each province, both in metropolitan and regional areas. In order to have sub-samples large enough to conduct separate statistical analyses, some types of contexts were oversampled (for example, the Quebec Anglophone colleges). However, even though the study did not aim for complete representativity, the female/male ratio is close to that of the colleges' and universities' populations, as well as that of Francophone Canadians, Anglophones Canadians, and first- and second-generation immigrants. (2)

Quebec Sample

The Quebec sample was comprised of 1,195 students recruited in six French (n=602) and four English (n=593) colleges. Among these, one English (n=473) and two French (n=319) colleges were in the greater Montreal area. Sixty-three percent of respondents were female and 37 percent male, and their mean age was 19. Fifty-six percent declared French as a first language, 24 percent English, and 8 percent both. Eight percent declared a different language, whereas 4 percent declared two or more languages, including at least one of the two official languages. Fifty-five percent were Francophone Canadians (born in Canada of parents both born in Canada), 15 percent Anglophone Canadians, 10 percent first-generation Canadians (born in Canada with at least one parent born outside Canada), and 20 percent second-generation Canadians (born outside of Canada of parents both born outside Canada).

New Brunswick Sample

In New Brunswick, the sample was comprised of 515 students recruited throughout the province in four French (n=187) and four English (n=328) community colleges and universities. Sixty-seven percent of respondents were female and 33 percent male, and their mean age was 18. Thirty-five percent declared French as a first language, 55 percent English, and 7 percent both. Three percent declared another language or two or more languages, including at least one of the official languages. Thirty-one percent were Francophone Canadians, 48 percent Anglophone Canadians, 12 percent second-generation Canadians, and 9 percent first-generation Canadians.

Alberta Sample

In Alberta, 823 respondents studying in five English community colleges and universities filled out the questionnaire. Seventy percent were female and 30 percent male, and their mean age was 20. Three percent declared French as a first language, 76 percent English, and 3 percent both. Eleven percent declared another language, whereas 7 percent declared two or more languages (including at least one of the two official languages). Two percent were Francophone Canadians, 56 percent Anglophone Canadians, 32 percent second-generation Canadians, and l0 percent first-generation Canadians.

Procedure

The various community colleges and universities were selected in order to include contexts in which the groups compared (Francophones, Anglophones, immigrant minorities) either had a majority or a minority status. Printed material was first sent to the selected institutions to inform them of the nature of the research project and to request their participation. A meeting was scheduled a few weeks later to present the project to the Director of Studies and to teachers, preferably in charge of mandatory subjects (French or English and philosophy) or, in some cases, of other social-science-related subjects (psychology, political science, sociology). The teacher or a member of the research team then presented the study in the classroom, informed students of the anonymity of their participation, and then invited them to fill out the questionnaire. When time allowed, students completed the questionnaire immediately, which ensured an almost one hundred percent response rate. In other cases, they were invited to fill it in during their free time and to hand it back to their teacher, which resulted in a response rate varying between fifty and seventy percent.

Instrument

Gagnon and Pagd's (1999) citizenship dimensions were used as the framework for the development of the questionnaire, resulting in a three-dimensional model merging national and cultural identities. A pre-test was first conducted among a sample of 300 college and first-year university students in Quebec, and factor analyses were performed to identify problematic items that were subsequently modified or eliminated from the final version of" the questionnaire (Page, Jodoin, and Chastenay 1999; Page, Chastenay, and Jodoin 1999). Factor analysis performed separately on the sections of the questionnaire corresponding to each of the three citizenship dimensions allowed the identification of different factors that provide an empirically-validated definition of the variables included in each of these dimensions (Page and Chastenay 2002). Some of the items used to assess these variables are presented in Table I.

The theoretical construct guiding this study was derived from the combination of both the literature review and the results of the factor analyses performed on the data in each of the three samples. Groups of intercorrelated items in the questionnaire were thus designated as scales measuring the different variables comprising each of the three main dimensions of citizenship. However, as these analyses were performed separately in the three provinces, although all but one of the variables emerged in all three samples, the items comprising some scales vary from one sample to the other, possibly revealing contextual variations (Page and Chastenay 2002).

Identity

An individual is thus characterized as a citizen by his/her relation to the main components of the surrounding sociopolitical reality: civic collectives of which s/he is a member; particular ethnocultural, national, and linguistic group(s) to which s/he belongs; the equality of all members of the society; and civic participation.

As a citizen living in Quebec, s/he has two principal civic identities: Quebecois (3 items, [alpha]=0.81) and Canadian (4 items, [alpha]=0.85), because s/he is counted as a member of the two civic collectives where s/he resides. This variable measures how identity is related to these collectives, which implies the sentiment of sharing certain characteristics with the other members of the collective, to feel a social proximity with them, and to link their collective self-esteem to the evaluation made of this collective by others. It is interesting to note that in New Brunswick and Alberta, only one factor of civic identity was found, combining items relating to the province and country in New Brunswick (5 items, [alpha]=0.87), as well as the city in Alberta (7 items, [alpha]=0.82). The results in these two last provinces confirm the Russian matriochkas theory, by which different levels of civic identity are conceived as being included within one another (Allen, Wilder, and Atkinson 1983; Salazar 1998). This suggests that Quebec presents a unique context where identification to the province and the country are statistically independent from one another and can be combined at all levels. In the questionnaire, the identity section included the city/town/village of residence as a third level of civic identity, as well as another sell-declared social group. However, preliminary analyses revealed that those two identifications varied very little across sub-samples and were thus not kept in the final analyses.

In the multinational and multicultural Canadian society, the individual citizen, by birth or through adoption, is a member of a sub-group identified by an ethnic ancestry, a national subgroup, or a particular linguistic community. This variable, designated as cultural identity (QC: 5 items, [alpha]=0.80; NB: 4 items, [alpha]=0.84; AB: 6 items, [alpha]=0.86), refers to the distinct groups to which students declare that they belong and to what extent they believe they share common characteristics with the other members of their group, feel a loyal obligation toward them, and link their self-esteem to the evaluation of this group.

Equality

Beyond the characteristics that distinguish them, the equality of citizens in a society is the principal norm of citizenship in modern democracies; the most significant aspect of citizenship is indeed its universality (Schnapper 1998). This dimension more specifically covers the extent to which young students agree with the application of the norm of equality to all their fellow citizens, even if their membership in a national or ethnocultural group distinguishes them from the majority or if their membership in the society is more recent. Five aspects emerged from the factor analyses and assess an egalitarian relationship to fellow citizens who are members of minorities.

The inclusion of diversity in the representation of the collective identity (3) of the provincial society is an aspect upon which equality theorists attach much importance, among them Charles Taylor, who talks about a "sharing of identity space" (1999) (QC: 19 items, [alpha]=0.89; NB: 14 items, [alpha]=0.91; AB: 18 items, [alpha]=0.92).

The attraction of social relationships with different people is an aspect that touches on the egalitarian relation to diversity From a more interpersonal angle without reference to specific types of differences (for example, gender, age, ethnicity, socioeconomic status, ability, etc.) (QC: 4 items, [alpha]=0.67; NB: 3 items, [alpha]=0.60; AB: 4 items, [alpha]=0.65).

The egalitarian relationship also fosters the presence of people of minority ethnocultural, religious, and linguistic identities in politics, in the management of firms and public services, and in all residential areas, in the presence of other languages on commercial posters, and in the manifestation of cultural diversity in all public places. This variable thus measures the attitude toward the presence of multiple cultural identities (QC: 12 items, [alpha]=0.79; NB: 6 items, [alpha]=0.69; AB: 6 items, [alpha]=0.72).

Canada and Quebec have promulgated charters of individual rights and freedoms. These charters protect minorities against any unequal treatment because of difference, and their application is often the object of judiciary decisions. These tribunal decisions are prescriptive, but the acceptance of the obligations they impose is subject to an individual evaluation, one that can be more or less positive, designated here as the attitude toward accommodations (QC: 3 items, [alpha]=0.84; NB: 6 items, [alpha]=0.81; AB: 7 items, [alpha]=0.83).

In Quebec, the Charter of the French Language decrees that French is "the language of the State and of the Law as well as the normal and usual language at work, at school, in communications, in commerce and in business" (Government of Quebec 1996, 13). But the Charter does not preclude the public usage of languages other than French, in particular English, because respect for minorities is one of the main principles of Quebec's linguistic policy. There exists, therefore, an undetermined space allowing citizens to use English as well as other languages when they wish, when they call, or when they judge it appropriate. The attitude measured toward the public usage of languages other than French (linguistic diversity) expresses the more or less positive evaluation of the use of this margin of freedom allowed by the Charter (QC: 4 items, [alpha]=0.90).

The items used to measure linguistic diversity vary in the other two provinces to reflect their specific legal and social linguistic realities. In the case of New Brunswick, the provincial government, in recognition of the linguistic composition of its population, decreed official bilingualism and adopted Bill 88 in 1981. The attitude measured is thus support for bilingualism (NB: 3 items, [alpha]=0.61). In Alberta, an officially Anglophone province, items in the questionnaire measure the extent to which languages other than English are welcome in the public sphere (AB: 3 items, [alpha]=0.65).

Participation

Finally, the relationship to civic participation is measured by three variables. Recent investment in student activities and community organizations is designated as current participation (QC: 13 items, [alpha]=0.77; NB: 5 items, [alpha]=0.77; AB: 5 items, [alpha]=0.76).

The will to invest in activities within communities, in politics, and in opinion or interest group activities is the intention for future participation (QC: 9 items, [alpha]=0.79; NB: 5 items, [alpha]=0.80; AB: 9 items, [alpha]=0.83).

The third variable measures the motivation to participate according to efficiency of participation (the evaluation of the impact citizens can have when they invest themselves, as individuals, in political life and in civil society) (QC: 6 items, [alpha]=0.66; NB: 6 items, [alpha]=0.67; AB: 5 items, [alpha]=0.68).

Analyses and Results

In order to answer these questions, second-order factor analyses were performed separately for each sub-sample in each province (Francophone Canadian, Anglophone Canadian, second- and first-generation Canadian), using the identity variables (three in the case of Quebec and two in New Brunswick and Alberta), the five equality variables, and the three participation variables, in order to see if a three dimensional factorial solution would be found, as suggested by the studies reviewed above. Tables 2, 3, and 4 present the mean and standard deviation of these variables for all sub-samples studied.

The principal-axis method of extraction was used (with oblimin rotation), testing for a three-factor solution. This method was chosen because it does not necessitate a strict normality of the distribution of items (McDonald 1985). As recommended, samples smaller than ten times the number of variables included in the factor analysis were excluded (Stevens 1996), notably the New Brunswick first- and second-generation Canadian sub-samples and the Alberta French Canadians. The first-generation Canadian sub-sample of the Alberta data was kept although it only amounts to eighty-five respondents, as well as the first-generation Canadian sub-sample of the Quebec data (n=105). Results for these two particular sub-samples should thus be taken as possibly less stable than the other sub-samples' results.

We also opted, when possible, for factorial solutions in which emerging factors were composed of at least three variables in order to have a statistically stable construct. Although we hypothesized for three-factor solutions, analyses revealed that this criterion resulted in two-factor solutions in the nine Canadian sub-samples.

Finally, it is important to note that in all Canadian sub-samples, two or more variables were excluded from the final factorial solution, having null or extremely low ([less than or equal to] 0.37) factor loadings (shown by gray zones in Tables 5 to 7, pp. 101, 103-107). These exclusions could suggest either that these variables are not part of the core dimensions of citizenship or that they are part of other dimensions of citizenship not covered by the questionnaire.

Quebec Results

As can be seen in Table 5, the Quebec results show two-factor solutions for all four sub-samples with variations between them. We first find that collective identity, which we believed represented an aspect of the support for equality norms, merges with the identity variables in all four sub-samples. Each sub-sample also excludes one or more variables, their covariation with other variables being limited, suggesting that the excluded variables are not related to the emerging factors. The fact that sub-samples do not exclude the same variables also reveals that what constitutes citizenship might vary, at least partially, according to the population studied.

We found a merging of the identity and participation variables for both the Francophone and Anglophone Canadians. However, Francophone Canadians exclude Canadian identity as well as current participation, whereas Anglophone Canadians exclude accommodations and linguistic diversity. The first factor merges identity and participation variables (Francophone eigenvalue: 1.9, 20.5% of variance explained; Anglophone eigenvalue: 2.1, 24%) and the second factor, in both sub-samples, includes the remaining equality variables (Francophone eigenvalue: 1.5, 16.5%; Anglophone eigenvalue: 1.1, 12%). It is interesting to note that one equality variable, linguistic diversity, is excluded by Anglophone Canadians and included by Francophone Canadians in the identity/participation factor, but is negatively loaded, thus indicating an inverted relation to the other variables comprising this factor.

The results are more complex for the other two sub-samples. In the first-generation Canadians' sub-sample, cultural and collective identities comprise the first factor along with accommodations, future participation, and efficiency (eigenvalue: 1.7, 22%). The second factor is comprised of Quebec identity, presence, and linguistic diversity (the last two loading negatively) (eigenvalue: 1.3, 16%).

Finally, for second-generation Canadians, the first factor includes Quebec (negative loading) and Canadian identity, as well as the three remaining equality variables after the exclusion of collective identity and attraction (eigenvalue: 2.0, 25%). The second factor is comprised of the three participation variables (eigenvalue: 1.3, 16%).

New Brunswick Results

In New Brunswick, as shown in Table 6, we also find a two-factor solution for both sub-samples. Francophone Canadians combine, in the first factor, civic and cultural identities with efficiency of participation (eigenvalue: 1.7, 25%). Linguistic diversity and both current and future participation were excluded from the analysis. This is the only sub-sample in which collective identity is included in the second, equality factor that includes the four remaining equality variables (eigenvalue: 1.1, 16%).

The same three variables, plus collective identity, are also excluded in the Anglophone Canadian sub-sample. The first factor combines civic and cultural identity with efficiency of participation (eigenvalue: 1.6, 26%). We find the remaining three equality variables in the second factor (eigenvalue: 1.2, 20.5%).

Alberta Results

We also found two-factor solutions in the three Alberta sub-samples (see Table 7). Here again, collective identity merges with identity variables in all three sub-samples.

Anglophone and second-generation Canadians share an identical factorial structure, and both of these sub-samples exclude current participation. The remaining four subdimensions of the equality dimension constitute the first factor (Anglophone eigenvalue: 2.1, 23%; second-generation eigenvalue: 1.4, 16%), whereas the second factor merges the identity and participation variables (Anglophone eigenvalue: 1.2, 14%; second-generation eigenvalue: 1.0, 11.5%).

Finally, in the first-generation Canadians' two-factor solution, linguistic diversity being excluded, a first factor includes the identity variables, current and future participation, and efficiency (eigenvalue: 2.1, 23%), while the second factor is comprised of the remaining three variables of the equality dimension (eigenvalue: 1.4, 16%).

THE BELGIAN STUDY

The Belgian study, conducted by Phalet and Swyngedouw (2001; 2002), studied actual representations of the citizenship of "ordinary" citizens.

Sample

The sample was comprised of non-national Moroccans and Turks compared to a Belgian sub-sample selected for its comparability in terms of socioeconomic background. The sample was composed of 404 Belgians (217 men, 187 women), 587 Turks (296 women, 291 men), and 391 Moroccans (157 women, 234 men).

The Turkish and Moroccan immigrant sub-samples were random samples from the population register, whereas the Belgian native comparison sub-sample was drawn selectively to match the age by gender by level of education structure of the immigrant sub-samples. Data were poststratified to correct for over- or under-representation due to selective non-response and to enhance the representativity of the Turkish and Moroccan populations in Brussels.

Procedure

Respondents participated in standardized, face-to-face personal interviews by same-sex, co-ethnic interviewers at the respondents' homes. Interviewers were minimally bilingual, and interviews were conducted in the respondent's language of choice. French, Dutch, Turkish, and Moroccan-Arab versions of the questionnaires were translated, backtranslated, and corrected for cultural bias by a multilingual committee. In addition, monolingual, Berber-speaking Moroccan respondents were matched with Berber-speaking interviewers who translated the questionnaire orally for them.

Instrument

In a questionnaire designed to study a vast range of variables (values, discrimination, individualism, languages, etc.; see Swyngedouw, Phalet, and Deschouwer 1999), respondents had to evaluate the importance of a list of eleven elements (ten in the case of Belgians) in Belgian culture and society comprising: symbols of identity (nationality, king, and flag); public institutions and services representing means and occasions of participation (languages, law, education, and employment); and rights and values corresponding to social norms (social security, freedom of expression, women's rights, and children's upbringing). The list of elements from social representations of Belgian and/or Turkish/Moroccan society and culture was rated on likert scales (from "1 = not important" to "5 = very important").

Analyses and Results

Factors were extracted across the three groups using separate principal components analyses with varimax rotation in the three sub-samples in order to highlight the specific configurations within each group as opposed to selecting only what was common across groups. Table 8 presents the elements' mean and standard deviation (the latter bolded and italicized) in the three sub-samples. As the purpose of the original study was to analyze transnational aspects of immigrant citizenship, similar elements were included in the questionnaire for the Turkish and Moroccan contexts. These results can be found in Phalet and Swyngedouw (2001; 2002) and Swyngedouw et al. (1999).

As can be seen in Table 9, for the Belgian sub-sample, the first (participative) factor includes languages, law, education, and employment (eigenvalue=2.37, 24% of variance explained). The second (identity) factor comprises nationality and the Belgian king and flag (eigenvalue=2.3, 23%). Finally, social security, freedom of speech, and women's rights form the third (normative) factor (eigenvalue=1.99, 20%).

The Turks share the same first, participative factor, to which they add social security and nationality (thus both identity and normative elements) (eigenvalue= 2.93, 27%). Nationality loads equally on the second (identity) factor, along with the Belgian king and flag (eigenvalue = 1.85, 17%). Finally, the third (normative) factor comprises freedom of speech, women's rights, and the Belgian method of child rearing (eigenvalue=1.6, 14.5%).

In the Moroccan sub-sample, the results reveal a more complex pattern of relationships. The first factor comprises the Belgian king and flag, women's rights, and Belgian upbringing, thus combining both identity and normative elements (eigenvalue=3.61, 33%). The second includes nationality, employment, freedom of speech, and languages (eigenvalue=1.92, 17.5%), which loads equally on the third factor, also comprised of law, education, and social security (eigenvalue=0.99, 9%). The second factor includes identity, normative, and participative elements, whereas the third is comprised of both normative and participative elements.

DISCUSSION

The main question asked in the introduction to this article is that of the relationship between the definitional elements of the concept of citizenship: identity, equality/norms, and participation. In the Canadian study, the three dimensions were divided into ten or eleven subdimensions, and second-order factor analysis showed how these variables were associated with one another to form a two-factor structure in all sub-samples. Similarly, factor analysis of the Belgian data allows us to see how twelve elements, theoretically measuring three similar dimensions, were associated to form three distinct factors in the three sub-samples.

Some principal conclusions can be derived from the data presented. The first to be drawn is that, in all but one sub-sample studied in Canada, subdimensions of identity and participation are associated in the same factor. This factor's composition varies from one sub-sample to the other (see Tables 5 to 7), but even if we observe certain differences between the specific identity and participation variables included, the link between these two dimensions is nevertheless not refuted.

A notable difference in the composition of the factor is the absence of the Canadian identity variable in the young Francophone Canadians of the Quebec sub-sample. The absence of a subdimension, following the logic of the analysis, does not indicate that this identity subdimension is very low or null in this sub-sample. On the contrary, this sub-sample's mean indicates that, in this case, Canadian identity is at mid-scale (4.57/10, s.d. 2.44). Rather, this implies that Canadian identity does not covary in the same direction as the other variables in this factor. On this basis, we must conclude that identification with Quebec and Canada's civic communities do not evolve in parallel for this sub-sample, nor are they antagonistic. On the contrary, the relation between the two can take any direction.

Subdimensions of participation vary very little from one sub-sample to another. In all cases, we find an association between the same two dimensions: future participation and efficiency of participation. In some cases, the factor also includes current participation, often correlated to future participation.

What is the nature of this association? The method of analysis demonstrates a descriptive association between subdimensions of identity and participation. Nothing allows us to conclude that this relation is unidirectional; it could very well be bi-directional, with identity and participation dimensions supporting one another. The stronger the civic identity, and in some cases particular cultural identities as well, the more the citizen is inclined to participate; the more s/he invests in participation, the stronger her/his identity. Theoretically, this relationship can be understood easily: citizens who feel a stronger membership in civic and cultural communities feel more concerned by topics of public deliberation and thus take part more readily in the decision-making process or get involved in civil enterprises. But one must not forget that the relation's hi-directionality can be interpreted otherwise: the citizen who gets interested in topics of public deliberation, perhaps because s/he seeks personal advantage, ends up developing stronger bonds with the civic and cultural community. However, this interpretation remains hypothetical. An experimental study in which both dimensions would be controlled hardly seems conceivable, but research using the Q-sort method might contribute to analyzing these relations with more precision.

A second general conclusion applies to most Canadian sub-samples wherein the equality subdimensions are associated and form a factor independent from the first. The composition of this equality factor also only varies a little. What does this tell us about the relationship between the dimensions? Results reveal that the equality factor is not correlated to the other factor. The agreement with norms of equality for all citizens, whatever their cultural membership, is a psychological disposition that seems to evolve independently from identity and participation. We can hypothesize that one or more psychological or social variables (e.g., socioeconomic status, personality, etc.) moderate the relationship between these dimensions, such that the latter might be positive for some subgroups and negative or neutral for others.

Thus, we cannot think that citizens showing a higher civic identification and a stronger proclivity to participate are also more favorably disposed toward cultural diversity, nor can we believe the reverse. Citizens with a more favorable disposition are not those who participate the most or the least (those who show the highest or lowest civic and/or cultural identity levels). On this basis, we can conclude that the strength of identities is neither favorable nor unfavorable to an egalitarian disposition among young adults. Public discourses that pretend that Canadian identity is one founded on the adherence to values embodied in the Canadian Charter of Rights and Freedoms are thus invalidated by such an observation. Similarly, those inclined to think that a strong Quebec identity founded on Francophone culture necessarily prevents a frank openness to cultural diversity are also invalidated. The identity dimension can thus be conceived as linked to a potential disposition toward cultural diversity, the link possibly being influenced by mediating factors that are, hypothetically, numerous (education regarding rights, a more favorable life experience with cultural diversity, etc.) and would certainly represent a fruitful research area.

The research conducted on three Belgian sub-samples by Phalet and Swyngedouw presents a somewhat different perspective. In this case, although not identical across the three sub-samples, the factorial solutions distinguish three dimensions relating to the national Belgian identity (identification to significant symbols of this identity), socioeconomic inclusion and participation, and, finally, social norms and equality of rights.

The nationality variable is particularly interesting as it holds a more complex position in the Moroccan and Turkish sub-samples. Whereas Belgian respondents associate nationality solely with the Belgian king and flag, in the second (identity) factor, the Turkish sub-sample reveals that nationality loads the same on the first (participation) factor as it does on the second (identity) factor. Even more complex, Moroccan respondents associate nationality with elements of participation (languages, employment) and norms (freedom of expression), whereas they combine, in another factor, symbols of identity (king and flag) with normative aspects (women's rights and children's upbringing).

Notwithstanding the fact that this second study covered a more restricted range of variables in a smaller sample, we find, in the Turkish sub-sample, that there is a cross-loading of an identity variable (nationality) with the participation factor, and, in the Moroccan sub-samples, that the three identity variables (nationality, Belgian king, and flag) are found in factors comprised of participative and normative variables, revealing complex interrelationships that vary from one sub-sample to another. As was the case in the comparison between participants of Canadian and immigrant origin in Quebec, we find that non-national citizens in Belgium seem to associate elements of citizenship differently than their national counterpart.

The Belgian and Turkish sub-samples share almost entirely identical factorial structures. It is first interesting to derive conclusions on the configuration itself and to interpret the meaning of the similarity between the two sub-samples. Must we conclude that, like native Belgians, Turkish immigrants have developed a Belgian national identity, a solid socioeconomic integration, and an evenly developed sense of equality of rights?

Whatever the case, results show that, among Turkish immigrants as well as native Belgians, these three dimensions can be distinguished from one another and that the participation dimension is the most important or valued, in line with a "social contract" type of citizenship (liberal rather than republican or communitarian) shared by natives and Turkish immigrants. There appears to be no link between the strength of national identity, socioeconomic inclusion, and a sense of equality; individuals can present very different profiles in this regard.

If we are able to interpret this similarity between native Belgians and Turks, it becomes interesting to consider the Moroccan sub-sample (which also presents three dimensions, but with a very specific configuration of elements, each of the three factors combining variables from at least two of the three main dimensions). In this sub-sample, contrary to the two others, two normative elements are found to be associated in the first lector with two elements of Belgian identity. Does this suggest a relationship between Belgian identity and the acceptance of equal rights for women and Belgian upbringing? If such a link were confirmed, it would certainly present interesting information about the relationship between these two dimensions. The second lector, in which employment is associated with freedom of expression, languages, and nationality, represents means of access to participation, which can in turn contribute to the empowerment of citizens. Finally, social security is found, with education, law, and languages, in the third lector, best described as reflecting the state's institutions (school, police), and where social security would be perceived through the relationship with social workers. As Phalei and Swyngedouw (2001) argue, these last two lectors might reflect the unique Moroccan experience of institutional discrimination in Belgium. Whereas the second factor is composed of elements that can be considered to represent social inclusion and access to equal rights and chances, the third lector might conversely convey perception of the state's authority. The opposition between rights and authority in those two factors is best illustrated in the cross-loading of the languages variable, which can be perceived as a means of power both for the state's authority and the individual citizen's access to democratic rights.

In conclusion, we find common tendencies, notably a distinct number of factors in Canada (2) and Belgium (3), as well as the possibility of a link between aspects of identity and participation, very clearly revealed in the Canadian sub-samples and also suggested by the Turkish and Moroccan sub-samples' data in the Belgian study. Variations are also revealed across sub-samples within each society, suggesting that the relationship among variables is more complex than one might wish for in a model of citizenship aspiring to universality. Although officials need to have access to a relatively clear definition of what citizenship encompasses, this implies a certain normalization of an ideal that might not reflect actual forms of citizenship present in the population. Such research can thus contribute to the development of public policies that reflect the needs associated with forms of citizenship experienced by ordinary citizens in their daily life.

ACKNOWLEDGEMENTS

The Canadian researchers received funding from SSHRC (GREAPE 1997-2000), Heritage Canada (1999-2000), and Immigration and Metropolis (1999-2002).

Table 1

Examples of Items for the Variables in the Canadian Study

Dimensions of
 Citizenship Variables Examples of Items

Identity Civic How much does this statement apply
 Identity to you?
 (5 to 7 items) I feel like I have a lot in common
 with all Quebecers/New
 Brunswickers/Albertans/Canadians.
 Cultural How much does this statement apply
 Identity to you?
 (4 to 6 items) I feel close to the members of my
 cultural group; we are alike.
Equality Collective If you had to describe Quebec
 Identity today, how important would it be
 (14 to 19 items) for you to include the following
 elements in your description?
 Cultural heritage from Great
 Britain; diversity of religious
 practices; etc.
 Attraction Do you agree or disagree with this
 (3 to 4 items) statement?
 In any society, the more people
 are different, the more social
 problems there are.
 Presence Do you agree or disagree with this
 (6 to 12 items) statement?
 Political parties should be
 composed of a larger number of
 people from cultural minorities.
 Accommodations Do you agree or disagree with this
 (3 to 7 items) statement?
 For schools and workplaces to
 adapt to the cultural and
 religious diversity present in our
 society, we have to ... adapt,
 upon request, the menu of the
 school cafeteria to the food
 restrictions of certain religions;
 make exceptions to the work
 schedule to allow certain members
 of religious minorities to
 practice their religion; etc.
 Linguistic Do you agree or disagree with this
 Diversity statement?
 (3 to 4 items) Although English and/or French is
 the official language of Quebec/
 New Brunswick/Alberta, we should
 allow a place for the diversity of
 language spoken in Quebec/New
 Brunswick/Alberia.
Participation Current In the last twelve months, have
 Participation you participated in the following
 (5 to 13 items) ativities?
 Given food or used goods to a
 community organization; signed a
 petition; etc.
 Future Do you intend to engage in any of
 Participation the following activities once you
 (5 to 9 items) have the opportunity?
 Contact a politician to express
 opinions; give to fund raisers for
 people in need, etc.
 Efficiency Do you agree or disagree with this
 of Participation statement?
 (5 to 6 items) Personal involvement in one's
 community is an efficient means of
 bringing about positive change.

Table 2

Mean and Standard Deviation of the Eleven Citizenship
Variables in Four Quebec Sub-samples *

 Dimensions Francophone
of Citizenship Canadian (n = 608)

 Identity Quebec Identity 6.32
 (3 items) 2.29
 Canadian Identity 4.57
 (4 items) 2.44
 Cultural Identity 6.66
 (5 items) 1.97
 Equality Collective Identity 4.66
 (19 items) 1.36
 Attraction 5.51
 (4 items) 2.07
 Presence 5.70
 (12 items) 1.41
 Accommodations 3.91
 (3 items) 2.65
 Linguistic Diversity 2.95
 (4 items) 2.53
Participation Current Participation 3.27
 (13 items) 1.90
 Future Participation 3.65
 (9 items) 1.78
 Efficiency of Participation 4.11
 (6 items) 1.19

 Dimensions Anglophone
of Citizenship Canadian (n = 170)

 Identity Quebec Identity 3.52
 (3 items) 2.32
 Canadian Identity 7.92
 (4 items) 2.00
 Cultural Identity 6.37
 (5 items) 2.34
 Equality Collective Identity 4.59
 (19 items) 1.57
 Attraction 5.80
 (4 items) 2.10
 Presence 7.28
 (12 items) 1.26
 Accommodations 6.05
 (3 items) 2.93
 Linguistic Diversity 8.46
 (4 items) 1.90
Participation Current Participation 3.55
 (13 items) 2.05
 Future Participation 3.97
 (9 items) 1.80
 Efficiency of Participation 4.08
 (6 items) 1.18

 Dimensions 1st Generation
of Citizenship Canadian (n = 105)

 Identity Quebec Identity 4.06
 (3 items) 2.75
 Canadian Identity 6.57
 (4 items) 2.38
 Cultural Identity 7.06
 (5 items) 2.24
 Equality Collective Identity 4.91
 (19 items) 1.55
 Attraction 5.64
 (4 items) 2.11
 Presence 6.91
 (12 items) 1.23
 Accommodations 6.24
 (3 items) 2.45
 Linguistic Diversity 7.00
 (4 items) 2.25
Participation Current Participation 3.15
 (13 items) 2.00
 Future Participation 4.08
 (9 items) 1.95
 Efficiency of Participation 4.26
 (6 items) 1.18

 Dimensions 2nd Generation
of Citizenship Canadian (n = 224)

 Identity Quebec Identity 3.79
 (3 items) 2.46
 Canadian Identity 6.68
 (4 items) 2.41
 Cultural Identity 6.91
 (5 items) 2.15
 Equality Collective Identity 4.86
 (19 items) 1.57
 Attraction 6.02
 (4 items) 2.01
 Presence 7.19
 (12 items) 1.33
 Accommodations 6.36
 (3 items) 2.64
 Linguistic Diversity 7.27
 (4 items) 2.83
Participation Current Participation 3.65
 (13 items) 1.92
 Future Participation 4.05
 (9 items) 1.77
 Efficiency of Participation 4.26
 (6 items) 1.28

* All scores have been reported on a scale ranging from 0 to 10

Table 3

Mean and Standard Deviation of the Ten Citizenship
Variables in Two New Brunswick Sub-samples *

Dimensions Francophone Anglophone
 of Canadian Canadian
Citizenship (n = 146) (n = 219)

Identity Civic Identity 8.00 7.92
 (5 items) 1.89 2.09
 Cultural Identity 7.01 5.99
 (4 items) 2.15 2.38
Equality Collective Identity 5.18 5.37
 (14 items) 2.00 2.12
 Attraction 6.13 5.83
 (3 items) 2.05 2.22
 Presence 7.14 7.49
 (6 items) 1.69 1.67
 Accommodations 5.47 6.25
 (6 items) 2.03 2.30
 Linguistic Diversity 2.63 4.46
 (3 items) 2.34 2.11
Participation Current Participation 6.33 6.50
 (5 items) 2.85 2.95
 Future Participation 3.05 3.03
 (5 items) 2.00 2.06
 Efficiency of Participation 5.97 6.13
 (6 items) 1.60 1.60

* All scores have been reported on a scale ranging from 0 to 10

Table 4

Mean and Standard Deviation of the Ten Citizenship
Variables in Three Alberta Sub-samples *

 Dimensions Anglophone
 of Canadian
 Citizenship (n = 476)

Identity Civic Identity 5.75
 (7 items) 1.87
 Cultural Identity 6.17
 (6 items) 2.10
Equality Collective Identity 4.56
 (18 items) 1.85
 Attraction 5.43
 (4 items) 1.92
 Presence 7.48
 (6 items) 1.80
 Accommodations 6.12
 (7 items) 2.15
 Linguistic Diversity 4.45
 (3 items) 2.42
Participation Current Participation 4.68
 (5 items) 3.03
 Future Participation 4.55
 (9 items) 1.78
 Efficiency of Participation 6.17
 (5 items) 1.58

 Dimensions 2nd Generation
 of Canadian
 Citizenship (n = 262)

Identity Civic Identity 5.35
 (7 items) 1.82
 Cultural Identity 6.64
 (6 items) 2.14
Equality Collective Identity 4.79
 (18 items) 1.85
 Attraction 5.34
 (4 items) 1.91
 Presence 7.64
 (6 items) 1.48
 Accommodations 6.12
 (7 items) 1.89
 Linguistic Diversity 4.54
 (3 items) 2.44
Participation Current Participation 4.61
 (5 items) 3.14
 Future Participation 4.52
 (9 items) 1.55
 Efficiency of Participation 6.07
 (5 items) 1.58

 Dimensions 1st Generation
 of Canadian
 Citizenship (n = 85)

Identity Civic Identity 4.88
 (7 items) 2.11
 Cultural Identity 6.96
 (6 items) 2.12
Equality Collective Identity 5.29
 (18 items) 2.06
 Attraction 5.44
 (4 items) 1.96
 Presence 6.38
 (6 items) 2.10
 Accommodations 4.47
 (7 items) 2.25
 Linguistic Diversity 7.52
 (3 items) 1.75
Participation Current Participation 4.12
 (5 items) 3.04
 Future Participation 4.54
 (9 items) 1.92
 Efficiency of Participation 6.26
 (5 items) 1.60

* All scores have been reported on a scale ranging from 0 to 10

Table 5

Factor Loading of the Eleven Citizenship Variables in Four Quebec
Sub-samples

 Dimensions Francophone
of Citizenship Canadian (n = 608)

 Factor 1 Factor 2

 Identity Quebec Identity 0.70 --
 (3 items)
 Canadian Identity [ ] [ ]
 (4 items)
 Cultural Identity 0.72 --
 (5 items)
 Equality Collective Identity 0.46 --
 (19 items)
 Attraction -- 0.57
 (4 items)
 Presence -- 0.80
 (12 items)
 Accommodations -- 0.55
 (3 items)
 Linguistic Diversity -0.42 --
 (4 items)
Participation Current Participation [ ] [ ]
 (13 items)
 Future Participation 0.39 --
 (9 items)
 Efficiency of Participation 0.43 --
 (6 items)

 Dimensions Anglophone
of Citizenship Canadian (n = 170)

 Factor 1 Factor 2

 Identity Quebec Identity 0.52 --
 (3 items)
 Canadian Identity 0.48 --
 (4 items)
 Cultural Identity 0.61 --
 (5 items)
 Equality Collective Identity 0.52 --
 (19 items)
 Attraction -- 0.81
 (4 items)
 Presence -- 0.53
 (12 items)
 Accommodations [ ] [ ]
 (3 items)
 Linguistic Diversity [ ] [ ]
 (4 items)
Participation Current Participation 0.45 --
 (13 items)
 Future Participation 0.67 --
 (9 items)
 Efficiency of Participation 0.53 --
 (6 items)

 Dimensions 1st Generation
of Citizenship Canadian (n = 105)

 Factor 1 Factor 2

 Identity Quebec Identity -- 0.72
 (3 items)
 Canadian Identity [ ] [ ]
 (4 items)
 Cultural Identity 0.43 --
 (5 items)
 Equality Collective Identity 0.51 --
 (19 items)
 Attraction [ ] [ ]
 (4 items)
 Presence -- -0.65
 (12 items)
 Accommodations 0.53 --
 (3 items)
 Linguistic Diversity -- -0.67
 (4 items)
Participation Current Participation [ ] [ ]
 (13 items)
 Future Participation 0.61 --
 (9 items)
 Efficiency of Participation 0.60 --
 (6 items)

 Dimensions 2nd Generation
of Citizenship Canadian (n = 224)

 Factor 1 Factor 2

 Identity Quebec Identity -0.54 --
 (3 items)
 Canadian Identity 0.38 --
 (4 items)
 Cultural Identity [ ] [ ]
 (5 items)
 Equality Collective Identity [ ] [ ]
 (19 items)
 Attraction [ ] [ ]
 (4 items)
 Presence 0.71 --
 (12 items)
 Accommodations 0.53 --
 (3 items)
 Linguistic Diversity 0.87 --
 (4 items)
Participation Current Participation -- 0.42
 (13 items)
 Future Participation -- 0.79
 (9 items)
 Efficiency of Participation -- 0.57
 (6 items)

Table 6

Factor Loading of the Ten Citizenship Variables in Two New Brunswick
Sub-samples

 Dimensions Francophone
of Citizenship Canadian (n=146)

 Factor 1 Factor 2

 Identity Civic Identity 0.73 --
 (5 items)
 Cultural Identity 0.74 --
 (4 items)
 Equality Collective Identity -- 0.44
 (14 items)
 Attraction -- 0.65
 (3 items)
 Presence -- 0.73
 (6 items)
 Accommodations -- 0.39
 (6 items)
 Linguistic Diversity [ ] [ ]
 (3 items)
Participation Current Participation [ ] [ ]
 (5 items)
 Future Participation [ ] [ ]
 (5 items)
 Efficiency of 0.44 --
 ParticiUation (6 items)

 Dimensions Anglophone
of Citizenship Canadian (n=219)

 Factor 1 Factor 2

 Identity Civic Identity 0.72 --
 (5 items)
 Cultural Identity 0.58 --
 (4 items)
 Equality Collective Identity [ ] [ ]
 (14 items)
 Attraction -- 0.54
 (3 items)
 Presence -- 0.93
 (6 items)
 Accommodations -- 0.40
 (6 items)
 Linguistic Diversity [ ] [ ]
 (3 items)
Participation Current Participation [ ] [ ]
 (5 items)
 Future Participation [ ] [ ]
 (5 items)
 Efficiency of 0.71 --
 ParticiUation (6 items)

Table 7

Factor Loading of the Ten Citizenship Variables in Three Alberta
Sub-samples

Dimensions Anglophone
 of Canadian
Citizenship (n=476)

 Factor 1 Factor 2
Identity
 Civic Identity -- 0.48
 (7 items)
 Cultural Identity -- 0.49
 (6 items)
Equality
 Collective Identity -- 0.48
 (18 items)
 Attraction 0.64 --
 (4 items)
 Presence 0.82 --
 (6 items)
 Accommodations 0.53 --
 (7 items)
 Linguistic Diversity 0.57 --
 (3 items)
Participation
 Current Participation [ ] [ ]
 (5 items)
 Future Participation -- 0.51
 (9 items)
 Efficiency of -- 0.58
 Participation (5 items)

Dimensions 2nd Generation
 of Canadian
Citizenship (n=262)

 Factor 1 Factor 2
Identity
 Civic Identity -- 0.47
 (7 items)
 Cultural Identity -- 0.41
 (6 items)
Equality
 Collective Identity -- 0.48
 (18 items)
 Attraction 0.48 --
 (4 items)
 Presence 0.77 --
 (6 items)
 Accommodations 0.48 --
 (7 items)
 Linguistic Diversity 0.45 --
 (3 items)
Participation
 Current Participation [ ] [ ]
 (5 items)
 Future Participation -- 0.42
 (9 items)
 Efficiency of -- 0.42
 Participation (5 items)

Dimensions 1st Generation
 of Canadian
Citizenship (n=85)

 Factor 1 Factor 2
Identity
 Civic Identity 0.56 --
 (7 items)
 Cultural Identity 0.51 --
 (6 items)
Equality
 Collective Identity 0.53 --
 (18 items)
 Attraction -- 0.78
 (4 items)
 Presence -- 0.75
 (6 items)
 Accommodations -- 0.46
 (7 items)
 Linguistic Diversity [ ] [ ]
 (3 items)
Participation
 Current Participation -0.42 --
 (5 items)
 Future Participation 0.70 --
 (9 items)
 Efficiency of 0.61 --
 Participation (5 items)

Table 8

Mean and Standard Deviation of the Eleven Citizenship
Variables in Three Belgian Sub-samples *

 Dimensions Belgian Turkish Moroccan
of Citizenship (n=404) (n=584) (n=392)

 Identity Nationality 4.6 5.2 4.5
 1.6 1.5 1.6
 King 4.9 4.8 4.9
 2.0 1.6 1.6
 Flag 4.5 4.9 4.4
 1.9 1.6 1.6
Participation Languages 5.4 5.7 6.0
 (French/Dutch) 1.2 0.9 1.1
 Law 4.9 5.6 5.8
 (Justice and Police) 1.2 1.1 1.2
 Education 5.7 5.8 5.6
 (Schools) 1.1 0.9 1.2
 Employment 5.7 6.0 5.8
 1.2 1.1 1.4
 Social Norms Social Security 5.8 6.1 6.4
 1.1 0.9 0.9
 Freedom 5.5 5.6 5.3
 of Expression 1.3 1.1 1.8
 Women's Rights 5.2 5.4 5.0
 1.5 1.2 1.8
 Children's [ ] 3.7 3.6
 Upbringing [ ] 1.6 1.6

* All scores have been reported on a scale ranging from 0 to 7

Table 9

Factor Loading of the Eleven Citizenship Variables in Three Belgian
Sub-samples

 Dimensions Belgian
of Citizenship (n=404)

 Factor Factor Factor
 1 2 3

Identity Nationality -- 0.79 --
 King -- 0.88 --
 Flag -- 0.88 --
Participation Languages 0.67 -- --
 (French/Dutch)
 Law 0.71 -- --
 (Justice and Police)
 Education 0.80 -- --
 (Schools)
 Employment 0.74 -- --
Social Norms Social Security -- -- 0.71
 Freedom of Expression -- -- 0.84
 Women's Rights -- -- 0.79
 Children's Upbringing [ ] [ ] [ ]

 Dimensions Turkish
of Citizenship (n=584)

 Factor Factor Factor
 1 2 3

Identity Nationality 0.42 0.45 --
 King -- 0.86 --
 Flag -- 0.84 --
Participation Languages 0.69 -- --
 (French/Dutch)
 Law 0.73 -- --
 (Justice and Police)
 Education 0.72 -- --
 (Schools)
 Employment 0.72 -- --
Social Norms Social Security 0.63 -- --
 Freedom of Expression -- -- 0.59
 Women's Rights -- -- 0.67
 Children's Upbringing -- -- 0.75

 Dimensions Moroccan
of Citizenship (n=392)

 Factor Factor Factor
 1 2 3

Identity Nationality -- 0.74 --
 King 0.81 -- --
 Flag 0.77 -- --
Participation Languages -- 0.52 0.53
 (French/Dutch)
 Law -- -- 0.70
 (Justice and Police)
 Education -- -- 0.74
 (Schools)
 Employment -- 0.72 --
Social Norms Social Security -- -- 0.68
 Freedom of Expression -- 0.75 --
 Women's Rights 0.68 -- --
 Children's Upbringing 0.67 -- --

NOTES

(1.) For example, depending on the specific sample studied, the Belgian variables were measured by a total of ten or eleven questions, whereas, depending on the province, the Canadian study used fifty-seven to eighty-two questions.

(2.) Their first language was identified by a specific question in this regard as well as two questions relative to their linguistic use with their parents and siblings.

(3.) Although this dimension is closely linked to the heritage dimension discussed in the Canadian Dimensions of Citizenship section and thus also considered to be a subdimension of identity, this specific scale mainly measures the inclusion of diversity in the representation of collective identity. It was therefore included in the equality dimension rather than the identity one.

REFERENCES

Allen, V.L., D.A. Wilder, and M.L. Atkinson. Multiple Group Membership and Social Identity. In Studies in Social Identity, edited by T.R. Sarbin and K.E. Scheibe, 92 115. New York: Praeger Publishers, 1983.

Bloemraad, I. Citizenship and Immigration: A Current Review. Journal of International Migration and Integration I, 1 (2000): 3-37.

Gagnon, F,. and M. Page. Conceptual Framework for the Analysis of Citizenship in Liberal Democracies. Volume I. Conceptual Framework and Analysis 01 Citizenship in Liberal Democracies, Volume 2. Approaches to Citizenship in Six Liberal Democracies. Reports submitted to the Department of Canadian Heritage. Ottawa: Department of Canadian Heritage, 1999.

Government of Quebec. Comite interministeriel sur la situation de la langue francaise (C1SLF). Le Francais Langue Comlnune, Enjeu de la Societe Quebecoise. Bilau de la situation de la langue francaise au Quebec en 1995. Report of the interdepartmental committee on the situation of the French language, 1996.

Kymlicka, W. Mulitcultural Citizenship. Oxford: Clarendon Press, 1995.

McDonald, R.P. Factor Analysis and Related Methods. Hillsdale: Lawrence Erlbaum, 1985.

Page, M. L'Education a la Citoyennete devant la Diversite des Conceptions de la Citoyennete. In L'Education a la Citoyennete, edited by M. Page, F. Ouellet, and L. Cortesao, 41-54. Sherbrooke: University of Sherbrooke, Editions du CRP, 2001.

Page, M.. and M.-H. Chastenay. Citizenship Profiles of Young Canadians. Canadian Diversity/e Canadienne 2, 1 (2003): 36-38.

--.Jetmes Citoyens du Quebec et du Noveau-Brunswick (Research report submitted to the Department of Canadian Heritage). Montreal: GREAPE, University of Montreal, 2002.

Page, M., M.-H. Chastenay, and M. Jodoin. Young Citizens Today: Views of Self and of Society (Research questionnaire). Montreal: (Research Group on Ethnicity and Adaptation to Pluralism in Education GREAPE), University of Montreal, 1999.

Page, M., M. Jodoin, and M.-H. Chastenay. Analyse Preliminaire des Profils de Citoyennete (Research report). Montreal: GREAPE, University of Montreal, 1999.

Phalet, K., and M. Swyngedouw. Les Representations Sociales de la Citoyennete et de la Nationalite: Une Comparaison entre Immigres Turcs et Marocains et Beiges Peu Scolarises 5. Bruxelles. Revue Internationale de Politique Comparee 8, 1 (2001): 109-33.

--. National Identities and Representations of Citizenship: A Comparison of Turks, Moroccans and Working-class Belgians in Brussels. Ethnicities 2, 1 (2002): 5-30.

Portes, A., L.E. Guamizo, and P. Landolt. Introduction: Pitfalls and Promises of an Emergent Research Field. Ethnic and Racial Studies 22, 2 (1999): 217-37.

Rawls, J. Political Liberalism. New York: Columbia University Press, 1993.

Salazar, J.M. Social Identity and National Identity. In Social Identity: International Perspectives, edited by S. Worchel, J.F. Morales, D. Paez, and J.-C. Deschamps, 114-23. London: Sage, 1998.

Schnapper, D. La Communaute des Citoyens. Paris: Gallimard, 1994.

Stevens, J. Applied Multivariate Statistics for the Social Sciences. 3rd ed. Hillsdale: Lawrence Erlbaum, 1996.

Swyngedouw, M., K. Phalet, and K. Deschouwer, Eds. Minderheden in Brussel: Sociopolitieke Houdingen en Gedragingen. Brussels: Vrije Universiteit Brussel Press, 1999.

Taylor, C. Democratic Exclusion (and its Remedies). In Citizenship, Diversity and Pluralism, edited by A.C. Cairns, J.C. Courtney, R McKinnon, H.J. Michelman, and D.E. Smith, 265 87. Montreal/Kingston: McGill-Queen's University Press, 1999.

Torney-Purta, J., R. Lehman, H. Oswald, and W. Schulz. Citizenship and Education in Twenty-Eight Countries. Civic Knowledge and Engagement at Age Fourteen. Amsterdam: IEA, 2001.